By that criterium, would not all empirical science be conjecture? If so, I think we simply use the word "conjecture" differently. In any case, I digress. I blame the shifting winds off the lake. It is. I guess the word "conjecture" might be too strong, but now we're arguing pointless semantics. Not that we had an argument of any substance before. Blaming shifting winds off the lake is one thing, but it's hard to believe that there are no shifting winds in other ballparks. Later today, if I have a chance, I will look into this. Honestly, it's probably not one thing. It's probably a combination of wind, day/night games, and other things. One would wonder if the effect existed if they restricted the data to pre WWII when most games were not played under the lights. I don't know much about statistics, but since all of the other teams do show a home field advantage by these numbers, couldn't the data for the Cubs just be an outlier, that has no other cause than randomness? I guess what I'm wondering is, does this study attempt to answer "does team X have a home field advantage", in which case it seems like there may not be a large enough sample size, or is it trying to find out whether or not a home field advantage exists period, in which case, the Cubs data point might just be a random outlier. My gut feeling is that if you are trying to determine if a specific park gives a team a home field advantage you need more data than 11 seasons provides. But I don't know anything.